{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9893390191897654,
  "eval_steps": 500,
  "global_step": 58,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.6596145629883,
      "epoch": 0.017057569296375266,
      "grad_norm": 0.20048722624778748,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.0001,
      "reward": 0.5323661006987095,
      "reward_std": 0.3604428209364414,
      "rewards/accuracy_reward": 0.5323661006987095,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.4511775970459,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.20257917046546936,
      "kl": 0.00012809038162231445,
      "learning_rate": 2.5e-06,
      "loss": -0.0001,
      "reward": 0.5292968964204192,
      "reward_std": 0.3622816801071167,
      "rewards/accuracy_reward": 0.5292968964204192,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.4633987426757,
      "epoch": 0.17057569296375266,
      "grad_norm": 4656877.0,
      "kl": 367.2123446941376,
      "learning_rate": 2.956412726139078e-06,
      "loss": 14.6391,
      "reward": 0.6254464566707612,
      "reward_std": 0.3107393844053149,
      "rewards/accuracy_reward": 0.6254464566707612,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.3234436035157,
      "epoch": 0.255863539445629,
      "grad_norm": 3.9253804683685303,
      "kl": 0.009595108032226563,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0005,
      "reward": 0.7332589611411094,
      "reward_std": 0.20636546667665243,
      "rewards/accuracy_reward": 0.7332589611411094,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.3687561035156,
      "epoch": 0.3411513859275053,
      "grad_norm": 85.55030059814453,
      "kl": 2.504261016845703,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.1002,
      "reward": 0.7540178909897804,
      "reward_std": 0.20728632267564534,
      "rewards/accuracy_reward": 0.7540178909897804,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1023.1982208251953,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.11913275718688965,
      "kl": 0.010443115234375,
      "learning_rate": 2.1156192081791355e-06,
      "loss": -0.0,
      "reward": 0.7078125305473805,
      "reward_std": 0.22495246771723032,
      "rewards/accuracy_reward": 0.7078125305473805,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1022.2207748413086,
      "epoch": 0.511727078891258,
      "grad_norm": 0.16609279811382294,
      "kl": 0.02086181640625,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0002,
      "reward": 0.6819196730852127,
      "reward_std": 0.23855062574148178,
      "rewards/accuracy_reward": 0.6819196730852127,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1021.6817230224609,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.13880808651447296,
      "kl": 0.11976318359375,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0045,
      "reward": 0.6767857484519482,
      "reward_std": 0.2631826866418123,
      "rewards/accuracy_reward": 0.6767857484519482,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1019.6558319091797,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.11709701269865036,
      "kl": 0.05513916015625,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0012,
      "reward": 0.721875037252903,
      "reward_std": 0.24268740862607957,
      "rewards/accuracy_reward": 0.721875037252903,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1019.2009307861329,
      "epoch": 0.767590618336887,
      "grad_norm": 0.1280747801065445,
      "kl": 0.06370849609375,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0015,
      "reward": 0.716294676065445,
      "reward_std": 0.24778253957629204,
      "rewards/accuracy_reward": 0.716294676065445,
      "rewards/format_reward": 0.0,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1019.0522689819336,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.12121888995170593,
      "kl": 0.073321533203125,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0026,
      "reward": 0.7066964566707611,
      "reward_std": 0.2454331662505865,
      "rewards/accuracy_reward": 0.7066964566707611,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1019.5757049560547,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.11665258556604385,
      "kl": 0.074462890625,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0031,
      "reward": 0.742857177555561,
      "reward_std": 0.22821373529732228,
      "rewards/accuracy_reward": 0.742857177555561,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 1019.2984924316406,
      "epoch": 0.9893390191897654,
      "kl": 0.07273356119791667,
      "reward": 0.7217262263099352,
      "reward_std": 0.23794425589342913,
      "rewards/accuracy_reward": 0.7217262263099352,
      "rewards/format_reward": 0.0,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 1.272181348064523,
      "train_runtime": 20557.9051,
      "train_samples_per_second": 0.365,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 5,
  "max_steps": 58,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}